Should Kamala Harris gamble on a Blue Florida?
Success is unlikely, but the payoff is high.
In our last Model Talk, we discussed 128 possible paths to the White House based on different combinations of who wins the seven major swing states. Some scenarios (like Donald Trump or Kamala Harris sweeping all seven states) happen fairly often in the 40,000 simulations our model runs, but others are really unlikely. Trump winning every swing state but Arizona? Surprising. Harris losing Pennsylvania and Michigan but winning Arizona, Georgia, and North Carolina? Really surprising. But what about Harris winning Florida?
Florida hasn’t really been on our radar, and for good reason. The once quintessential swing state just isn’t anymore.1 Donald Trump won the state in 2016 and doubled his margin of victory in 2020, Republicans have supermajorities in both state legislative chambers, and the state hasn’t elected a Democratic senator since 2012. Does Florida’s rightward shift make the state a lost cause for Harris that isn’t worth an investment of her campaign’s limited resources? As usual, it’s not that simple.
The state of the race in the Sunshine State
As of today, our model gives Kamala Harris a 21 percent chance of winning Florida’s 30 electoral votes. Now, a one-in-five chance isn’t bad, but it isn’t great either. For perspective, this scenario is about as likely as Harris winning Alaska. So let’s be clear: Harris is the definite underdog here. There are however a few things that make the race more interesting than the 21 percent win probability would imply.
First, the Florida polls have slowly tightened since Harris entered the race. On the day our Harris-Trump model launched, Trump led by a 7-point margin in the state. Now? He only leads by 3.3 points. Harris has gotten some good individual polls recently too. A Victory Insights poll had her down by 2 points, and she trailed by 1 point a Bullfinch Group poll. Are these particularly high-quality pollsters? No. And she’s had some bad polls too. She trailed by 4 points in this Public Policy Polling survey — a firm that typically leans towards Democrats, and the poll was for a partisan client.
Trailing by around three and a half points isn’t great but it's a lot better than trailing by 7 points. At this point, a normal sized polling error that favors Harris could result in a Democratic victory. Now, it’s important to note that the model projects Trump to win the state by 3.7 points, slightly more than his 3.3 point margin in the polling average. In other words it doesn’t totally buy the polls and expects Trump to do a little better than what they currently indicate.
What else makes the Florida race interesting? Well, we can think of 30 things: it’s worth 30 electoral votes. That’s twice as many as Michigan and more than Pennsylvania and Wisconsin combined. It’s also why our model thinks Florida is the 7th-likeliest tipping point state — ahead of states like Nevada where the race is much more even but the number of electoral votes up for grabs is smaller.